@JanuaryChill
Why can't they spell out what Save the NHS was always about - if you /your dad/grampy/girlfriend gets seriously ill with Covid OR is in a bad car accident and ends up needing intensive care, be aware the nurses won't be able to pay as much attention to you/him/her as they usually would. Alarms on machines might go off and no one comes running immediately because there is no one at that second....
The problem is, the argument is always made that people may not care about getting COVID all that much, because they're low risk, but they should care about the NHS collapsing because that affects us all and so them directly - but that's only true to an extent. People at the lowest personal risk from COVID are also mostly the lowest users of the NHS (if we set aside small children, who aren't making their own choices on compliance). There's a reason the example given is so often 'if you're in a car crash', because that could happen to anyone - but people know it isn't terribly likely to happen to them. People do not generally organise their lives around the possibility of them having a bad car accident (you can tell this from how they drive).
I am personally extremely invested in hospital capacity - I'm going to give birth sometime in the next few weeks, I have a toddler who has needed A&E a couple of times. I also had suspected cancer that required exploratory surgery in 2019 and received amazing and almost frighteningly swift care from the NHS for this, so the 'what if you found a lump' argument also resonates with me. But when I was 25, say, I had never actually been in an A&E and I hadn't even seen a GP for anything but repeat pill prescriptions for years. Now, I think and hope I would still have wanted to protect the NHS then for the good of others/society, but it would have been hard to convince me that my personal risk was high.
So whether you're emphasising COVID itself or the risk of NHS collapse then the least at risk demographics overlap, and your options with them are either to emphasise altruism (don't kill granny/don't let grandad die unnecessarily of a heart attack because they're out of beds) or emphasising that the risk to them is real even if remote (people in their 20s have died of COVID/you could be in a car crash). And my guess is that neither is as effective as the messaging is for people who feel that they are actually personally at significant risk.